Astronomy & Astrophysics manuscript no. Lefever © ESO 2018 August 21, 2018 Statistical properties of a sample of periodically variable B-type supergiants ⋆ Evidence for opacity-driven gravity-mode oscillations K. Lefever1, J. Puls2, and C. Aerts1,3 1 Instituut voor Sterrenkunde, K.U.Leuven, Celestijnenlaan 200B, B–3001 Leuven, Belgium 2 Universit¨atssternwarte M¨unchen, Scheinerstr. 1, D-81679 M¨unchen, Germany 3 Departement Astrofysica, Radboud Universiteit van Nijmegen, PO Box 9010, 6500 GL Nijmegen, the Netherlands Received 14 July 2006 / Accepted 3 November 2006 ABSTRACT Aims. We have studied a sample of 28 periodically variable B-type supergiants selected from the HIPPARCOS mission and 12 comparison stars covering the whole B-type spectral range. Our goal is to test if their variability is compatible with opacity-driven non-radial oscillations. Methods. We have used the NLTE atmosphere code FASTWIND to derive the atmospheric and wind parameters of the complete sample through line profile fitting. We applied the method to selected H, He and Si line profiles, measured with the high resolution CES spectrograph attached to the ESO CAT telescope in La Silla, Chile. Results. From the location of the stars in the (log Teff , log g) diagram, we suggest that variability of our sample supergiants is indeed due to the gravity modes resulting from the opacity mechanism. We find nine of the comparison stars to be periodically variable as well, and suggest them to be new α Cyg variables. We find marginal evidence of a correlation between the amplitude of the photometric variability and the wind density. We investigate the Wind Momentum Luminosity Relation for the whole range of B spectral type supergiants, and find that the later types (> B5) perfectly follow the relation for A supergiants. Additionally, we provide a new spectral type - Teff calibration for B supergiants. Conclusions. Our results imply the possibility to probe internal structure models of massive stars of spectral type B through seismic tuning of gravity modes. Key words. Stars: atmospheres – Stars: early-type – Stars: fundamental parameters – Stars: mass loss – Stars: oscillations – Stars: variables: general 1. Introduction variability, with very diverse causes, within that sample. The study of Waelkens et al. (1998) is quite different in this respect, One of the remarkable by-products of the ESA HIPPARCOS arXiv:astro-ph/0611484v1 15 Nov 2006 as they came up with a sample of 29 new variable B supergiants mission was the discovery of a large amount of new periodi- exhibiting clear periodic microvariations at relatively short pe- cally variable B stars, including almost a hundred new slowly riods of one to a few days. The current paper contains a follow- pulsating B stars (SPBs hereafter) and 29 periodically variable up study of the latter sample. Our goal is to evaluatethe sugges- B-type supergiants (Waelkens et al. 1998; Aerts et al. 1999; tion by Waelkens et al. (1998) that these periodically variable Mathias et al. 2001). The photometric behaviour of different B-type supergiants experience oscillations excited by the opac- kinds of evolved massive stars were analysed in detail from ity mechanism, in analogy to main sequence B stars. HIPPARCOS data. VanLeeuwen et al. (1998) performed an analysis of 24 known B- to G-type variable supergiants and The suggestion of occurrence of non-radial oscillation found periods of tens of days for each of them, in agreement modes in massive supergiants is not new. Microvariations with with previous ground-based studies. Marchenko et al. (1998), amplitudes between a hundredth and a tenth of a magnitude on the other hand, focused on the HIPPARCOS variability of in the visual, and periods ranging from some 5 to 100d, have 141 O-type and WR stars and noted the remarkable variety of been found in numerous supergiants of spectral type OBA, termed α Cyg variables (Sterken 1977, 1983; Burki et al. 1978; Send offprint requests to: K. Lefever, e-mail: karolien@ster. van Genderen et al. 1989; Lamers et al. 1998; van Genderen kuleuven.be 2001). Burki (1978) considered a sample of 32 B- to G-type ⋆ Figures of the spectral line fits and discussion of the supergiants and derived an empirical semi-period-luminosity- individual objects are only available in electronic form via colour (PLC) relation (see his Eq.5), from which he suggested http://www.edpsciences.org the variability to be caused by oscillations rather than mass 2 Lefever et al.: Periodically variable B supergiants loss. Lovyet al. (1984) indeed managed to come up with a luminosity, mass-loss, rotational velocity, etc.) of the stars in theoretical PLC relation in agreement with the observed one the sample by Waelkens et al. (1998) using high-quality spec- for this type of stars. However, only 40% of the variable su- troscopic data. From this, we derive their position in the HR pergiants have periods compatible with the radial fundamen- diagram with high precision and check if the stars lie within tal mode, while the majority must exhibit a different kind of the instability strips of gravity mode oscillations predicted by oscillation mode. Kaufer et al. (1997) made an extensive high- Pamyatnykh (1999) and Saio et al. (2006). During our analy- resolution spectroscopic study of 6 BA-type supergiants which sis we also look for asymmetries in the metallic line profiles. they monitored for several years. They concluded that the vari- Further, we search for correlationsbetween the physical param- ability patterns are extremely complex and point towards cycli- eters and the photometric variability. In particular, we investi- cal variations in the radial velocities. From CLEANing the pe- gate if there is any connection between the observed peak-to- riodograms of the radial-velocity curves, they derived multiple peak amplitude and frequency of the light variability, the wind periods and assigned them to non-radial oscillations because density and the rotation of the stars. Finally, we investigate the the travelling features in the dynamical spectra turned out to be Wind momentum Luminosity Relation (WLR) for the complete incompatible with the rotational periods of the stars. B-type spectral range. Glatzel & Kiriakidis (1993) interpreted the periodic vari- ability of supergiants with masses above 40 M⊙ in terms of 2. Observations and Data Reduction strange-mode instabilities and showed the classical opacity mechanism to be unimportant in such objects. Glatzel et al. We selected all southernstars of luminosity class I or II brighter (1999) subsequently performed detailed calculations for stars than 9th mag in the sample by Waelkens et al. (1998) which with M = 64 M⊙ and predicted irregular velocity and luminos- fitted the observation window of the assigned telescope time. ity variations with time scales of 1 to 20 d. They also proposed This concerns 21 stars. To this we added 10 more candi- that such massive stars undergo pulsationally-driven mass-loss. date α Cyg variables from the Catalogue of Periodic Variables It is not clear how their result will change for stars in the mass of HIPPARCOS (ESA 1997), in such way that the complete rangeof 10 to 30 M⊙, which is the transition region from low to sample fully covers the B-type spectral range. Waelkens et al. large mass-loss rates due to line driving. Therefore,the periodic (1998) were unable to assign them to one of the five consid- variations of the B supergiants found by Waelkens et al. (1998) ered classes (β Cep stars, SPBs, chemically peculiar stars, B- might still be due to the classical opacity mechanism, since the type supergiants, Be stars), see also Kestens (1998). Our period instability strips of the β Cep stars and the SPBs were found to analysis and their spectral type clearly point towards α Cyg extend to higher luminosities shortly after the discovery paper variables, however. (Pamyatnykh 1999; Balona & Dziembowski 1999). These 31 targets (for spectral types, see Table6 and dis- Waelkens et al. (1998) positioned the new periodic B su- cussion in Sect.7) were added to the long-term spectroscopic pergiants in the HR diagram on the basis of multicolour photo- monitoringprogrammeof periodic B stars conductedat Leuven metric calibrations (accurate parallaxes are not available) and University (Aerts et al. 1999). The spectra of the stars were found them to be situated between the SPBs and previously gathered with the CES spectrograph attached to the CAT tele- known α Cyg-type variables (see their Fig. 2). Oscillations scope at La Silla during numerous observing runs spread over were not predicted in that part of the HR diagram. Given the two years. For most targets, we obtained two exposures of the uncertainty in the effective temperature and luminosity, Aerts Hα line in different seasons (in order to check for its variabil- (2000a) tried to improve upon the fundamental parameter de- ity), one of the Hγ line, oneof theHe I6678line and oneof the termination by constructing the spectral energy distribution of He I4471 line. Besides these, we observed one silicon line for the stars as a better diagnostic to derive their effective tem- each star with the goal to obtain an accurate temperature esti- perature and gravity. This did not improve the large uncer- mate. Depending on spectral type, this is either the Si II4130 tainties on the latter parameters, however. Nevertheless, the doublet (late B-type) or the Si III4560 triplet (early B-type). sample selected by Waelkens et al. (1998) remained the most The spectra were reduced in the standard way, i.e., by valuable one to investigate observationally the occurrence of means of flatfielding, wavelength calibration through Th-Ar gravity modes in supergiant stars, because it is unbiased in the exposures and rectification of the continuum through a cubic sense that the stars were not at all selected to be observed with spline fit.